Your browser is not supported. Please update it.

24 January 2007

Dazed and confused – Mobile TV faces tough decisions at Paris show

Right across Europe there remains no concise consensus about the future of mobile TV, if the speakers at the Mobile TV World Congress, held in Paris last week, were anything to go on.

Less of this conference was lent to the religious wars that have accompanied mobile TV for the past five years, you know DVB-H, versus T-DMB, versus MediaFLO versus Multicasting, but there remains confusion in the market with many speakers still convinced that unicast TV services were not only viable going forwards, but would remain dominant.

One speaker, from French aggregator Mobibase even went as far as saying that all the investment on mobile TV could be as wasted, as all the money spent on WAP was wasted, and then went on to show an orgy of mobisode short form content that made Faultline think we had gone through a ten year time warp back to the beginnings of unicast TV.

But for the most part there was sanity as operators moaned about the costs of building OFDM networks, but accepted their necessity, regardless of whether they were MediaFLO, DVB-H or DVB-SH, and pushed steadfastly towards strategies where dominant operators ‘teamed’ up on building out DVB-H, a strategy already employed on one system in Italy (Mediset) and which is set to be copied in France, Germany and Spain and even the UK if it ever finds any spare UHF spectrum.

Among all of this Nokia continued talking about the right way forward being through open standards, and then speaker Harri Mannisto, director of new Watch N series multimedia devices, confessed that Nokia had learned its lesson and would not be trying to railroad the community into any standards where it had an advantage, including the OMA broadcast standard for DRM, something it has been rounded on in the past for promoting, when the majority of players wanted to go down the DVB open framework approach.

Even Qualcomm, speaking up for MediaFLO through the FLOForum, highlighted this as a change. There should be DRM choice and it had opened its forum to DRM players and dropped the approach of its own OMA based, proprietary DRM. But MediaFLO continued to snipe at DVB-H through Kamil Grajski of the FLO Forum, a Qualcomm employee, that kept pointing out that networks would be cheaper to build if only Europe would embrace MediaFLO, which is so much more spectrally efficient, at least in his view.

This view of MediaFLO’s advantages over DVB-H are not given much credibility here, and on the back of broadcaster companies saying that the two systems are comparable on a dollar per bit basis. When you add this to the idea that DVB-T free to air terrestrial systems have an in-built cost mitigation because they offer DVB-H systems a large existing antenna and cabling and power sources and numerous other advantages, then DVB-H is going to be dominant in Europe and at least somewhat effective outside of it. But that doesn’t mean that MediaFLO won’t also find its rightful home.

About the most significant announcement at the show, if we don’t count the free to air DVB-H that 3 Austria is launching (See separate story) was the brave move by Orange France of what it is calling its HD Mobile TV.

Erwan Launay, a spokesman for Orange who sits on the 3GPP committee, suggested that there was absolutely no hurry to embrace datacasting such as DVB-H and talked about what the company calls HD Mobile TV, where its existing HSDPA customers, which Orange markets as 3G+, can already have unicast 250 kbps video augmented by 48Kbps audio, pretty much in line with DVB-H in that it is Quarter VGA density (320 x 240 pixels) at 30 frames a second.

Launay accepted that DVB-H would still be an improvement and said yes Orange is interested in that, but only because of things like the 10 seconds it currently takes to move back to its video portal on the handset and select a new channel, and the clunky way the interface works. He reminded the audience that Orange already has 55 TV channels and 2,500 VoD videos for download and was content to see DVB-H away into 2008.

While Launay accepted that only 30 people can watch unicast mobile TV from even an HSDPA base station, and that this would limit concurrent viewers over all of France to something under 1 million, he said that since the market was no-where near this level, Orange forsees no problems.

Which is why perhaps he was blanching at the €310 million ($370 million) price tag that one analyst, Eden Zoller from Ovum, was placing on the cost of building out a DVB-H network in France, since Launay clearly believes that mobile TV is not yet ready for the types of customer numbers that will pay back that kind of investment.

‘And you cannot pay for it with advertising, because if adverts were sold at the same price as terrestrial TV advertising, they would grab just 1% of the €4.4 billion ($5.2 billion) TV advertising market in France.’

Even so he is predicting a gradually move to both DVB-H and Multicasting using MBMS in time, and Orange covering any move that any other French operator makes.

Which is perhaps why Alcatel, thought of as a ‘best of friends’ supplier to with Orange’s parent, France Telecom, was keen to point out that there are more than two ways to skin a cat and came up with a €90 million ($100 million) price tag to bring mobile TV to France using satellite based DVB-SH in S-Band.

Alcatel had wheeled out their big wheels in Olivier Coste, President of Alcatel Lucent Mobile Broadcast to deliver this message, and he said, ‘A network across France that uses DVB-SH satellite to reach 60% of the population and which has 3,600 terrestrial gap fillers in all of the towns, would cost just €90 million to build. This would deliver 45 TV channels at a total cost of just €2 million ($2.4 million) per channel per year.’

Presumably the Opex costs include all the satellite costs and the costs of looking after all those gap filler transmitters.

Coste also let out the fact that 28 member states in Europe had accepted the Alcatel proposal to use S-Band spectrum in this way and that formal approval would be through in a matter of months.

In this way a free to air service with premium channels could be set up, said Coste, and not at the €1 ARPU improvement that passive national TV advertising would give operators, but by incorporating €10 of ARPU in addressable and local advertising.

Coste was impressed with the US experience where satellite radio has managed to get people to spend $13 a month on entertainment and yet reached 14 million people, but felt that this was a ceiling for similar services in Europe and that they only dreamt of getting that amount of money from such a service as a subscription.

Interestingly another speaker showed that the first German experiment with mobile TV, using T-DMB, offered by Debitel, shows just how unsuccessful T-DMB was going to be in Europe and revealed that it has only attracted 3,000 customers in the six months it has been on the market. Clearly operators at the conference were terrified that they would launch an expensive service, costing $100s of millions, only to find a similar number of customers willing to buy into that service.

Finally a speaker from the French Mobile TV Forum talked about initiatives progressing through the French government and a decision about a UHF slice of spectrum to come by March. When he also said that from there to having a DVB-H multiplex up and working would be possible by the Rugby World Cup, in September, he drew hoots of derision. Faultline’s discussions with implementers would suggest that even the combined might of the top three cellcos and top three French broadcasters, acting together, would be hard pressed to pull off such a feat, including building a network and sourcing, testing and delivering handsets, in such a short space of time.

As a final note one speaker who works with KDDI confirmed that the Japanese operator was not only experimenting with MediaFLO, but was also trying out multicasting using BCMCS, the 3GPP2 version of MBMS. So despite the established success of mobile TV in Japan through ISDB-T, where each mobile channel is integrated in one segment of the signal of a terrestrial TV service, its weakness on in-building penetration and on data rate is still giving rise to more mobile TV experiments there.